1. Field of the Invention
The present invention relates generally to the field of power generation and in particular to an improved interconnect for a solid oxide fuel cell.
2. Background of the Invention
Global demand for power generation in the next twenty years is expected to increase by about 2 million MW, of which 490,000 MW are projected to be powered by natural gas. Utility deregulation in the United States, concerns over health issues and capital costs associated with the transmission and distribution of electrical power make it likely that at least 30% of this natural gas fired capacity will be provided by modular power plants located in close proximity to the end users.
Solid oxide fuel cells are an attractive solution for meeting those needs for distributed power in a manner which is both energy efficient and environmentally sound. Solid oxide fuel cells offer modularity as well as higher fuel efficiency, lower emissions, and less noise and vibration than gas turbines or diesel generators. Data from test modules show that No.sub.x production is greatly reduced and almost non-existent in fuel cells. At the same time, fuel cell test modules have been tested to operate at greater than 50% efficiency.
In order to be widely accepted by delivering energy efficiently and in an environmentally sound manner, solid oxide fuel cells must be able to cost-effectively produce electricity and heat. The capital and operating costs of solid oxide fuel cells must compare favorably with alternative sources for distributed power, such as internal combustion engines and gas turbines.
Interconnect functionality and cost are two of the biggest barriers to producing market competitive solid oxide fuel cell generators. The interconnect must provide reactant gas separation and containment, mechanical support to the cells and a low resistance path for current connecting the cells electrically in series and/or in parallel. Meeting these functional requirements remains a challenge. Monolithic interconnects made of lanthanum chromite and high chromium alloys have been used with some success. However, both types are quite expensive and compromise aspects of the interconnect function.
Lanthanum chromite and high chromium alloys are currently cost prohibitive for use in commercial products with a conventional monolithic interconnect design. Projected costs, assuming high production volumes using net shape ceramic processing or a metal forming process, are potentially low enough to enable marginally cost competitive solid oxide fuel cell power generation. However, the gap between required startup cost and initial market size is a decisive barrier to solid oxide fuel cell commercialization.
Gas separation requires a dense impermeable material which does not have significant ionic conductivity. Alloy interconnects that have been developed readily satisfy this requirement. Ceramic processing has developed the capability to produce interconnects of sufficiently high density, however, many compositions have unacceptably high ionic conductivity. The known compositions of such ceramics possessing low ionic conductivity also have less than acceptable electronic conductivity or are not well matched to the coefficient of thermal expansion (CTE) of the cell.
Matching cell and interconnect coefficients of thermal expansion allows sealing of cells to interconnects for gas containment. Alloy interconnects generally have a higher CTE than the CTE of the cell. While the CTE of ceramic interconnects are more nearly matched than alloy interconnects, they are still lower than that of the cell. As a result, regions of the cell may be adversely displaced wherein it becomes difficult to effectively confine reactant gases to their intended flow paths, which in turn adversely affects the stack efficiency. While changes between room and operating temperatures produce the largest thermal displacements, temperature changes in a stack as reactant and current flows are varied can also create undesirable detrimental displacements.
Dissimilar thermal expansion characteristics also cause the relative motion imparted by thermal expansion to disrupt the electrical current path between the electrodes and interconnects. The contact resistance generated in this way significantly reduces stack performance and efficiency. In the case of alloy interconnects, the motion can dislodge a protective oxide scale and expose underlying unprotected material. Oxidation of the unprotected material increases the overall scale thickness, and as scale conductivity is comparatively poor, scale growth contributes directly to performance degradation.
The issues presented by oxide scale conductivity and growth are some of the most challenging of all those confronting developers of metal interconnects. Scale resistance is a function of oxide conductivity, thickness and continuity. Porous or laminar scales have the effect of increasing the current path length while reducing the effective current carrying cross sectional area. The mechanism for scale conductivity and growth are related such that scale growth rate increases with scale conductivity. Higher growth rates generally produce less dense, less adherent scales. Any alloy (other than noble or semi-noble metals) will have to compromise scale conductivity in order to control degradation due to scale growth. Coating the interconnect with a conductive oxide layer provides more control of the scale composition and microstructure but does not change the basic nature of the problem.
Thus, it is an object of the present invention to provide an interconnect for a solid oxide fuel cell which permits substantial matching of cell and interconnect coefficients of thermal expansion.
It is a further object of the invention to provide an interconnect region manufactured using vias to fill the interconnect space between the cell anode and cathode to match the material coefficients of thermal expansion.
It is also an object of the invention to separate the interconnect functions of gas separation and containment, from the current carrying function of the interconnect, thereby enabling selection of materials best suited to each function and atmosphere.